artificial-intelligence, machine-learning, deep-learning
Deep Learning (DL) is a type of machine learning and training to build a learner model of a large amount of data. This type of algorithm (DL) is built to learn "Learning Features" without having to define those properties in advance. In addition, it is one of the best algorithms that enable the machine to learn different levels of data characteristics (eg images).
In my view, deep learning has excelled in creating new characteristics that can be learned at different levels, and this may lead researchers in the future to focus on this important aspect.
Features are the first factor in the success of any intelligent algorithm in machine learning. Your ability to extract and / or select properties correctly, and represent and configure data for learning is the point between success and failure of the algorithm.
When we look at deep learning algorithms, we find that almost identical algorithms are in Neural Networks, abbreviated by NN. This is true to some extent. The only difference between these two types, NN and DL, is the depth of this algorithm or the amount of Hidden Layers in the algorithm.
Deep learning algorithms are therefore a natural extension of neural network algorithms. However, the increase in the number of hidden layers in deep learning compared to neural networks has complicated the learning in deep learning.